Wetlook World Forum

Message # 100861.4.1

Subject: Thank you!

Date: Tue 07/01/25 15:25:17 GMT

Name: Hudson

Email:

Website:

By Oldbewet - Mon 06/01/25 22:49:09 GMT

Website:

Picture No 1, the girl in the street leaning on lamppost: The wetness of her clothes is fantastic, seems to me very realistic, there are parts that are realy shine and other "just" clink to the body, even her hair look very realistic. little raindrops in the air, and the waterdrop on her face, i am fascinated. I have to admit, i even like the expression on her face. But just as you might expect, Eagle Eye is watching: There is an issue with her hand: the fingers seem a bit short, there no

structure on back of her hand, and compared to the skin color of the face und her belly the skincolor of her hand  does not really match, to me color of hand should be more similar to the face than to the belly.    And there is a little thing in the background that is disturbing me:The other side of the road looks like traffic light, but the green light seem not to match the carriage of the other lights.  Yes, it's just little detail, but somehow it draws my attention. Over all a very good picture. 9 of 10 points.

Picture No2: If spent quite some time looking for the usual errors in AI pictures, well i can't find any, the hair on top of her head look a bit strange, but not really unreal. Even the dirt on tile joint on top right corner of bathtub looks realitic. Overall not such a spektucular picture  as no1.And as it avoids all the points where AI normally fails  it score 9 of 10

Picture No3 , grey Top, brown leggins: beautiful girl, the wetness looks good, the splash of water at her feet looks better than any AI i have ever seen. But there seem to be a real issue with her arm and hands. The lenght of the arms remind of an orangutan, even her hands seem a bit long. 8 of 10  

Picture No4, green top with jeans: almost the same as picture No3: It seems to me, that on both hands a finger is missing. I guess in world without AI, no one would have looked for such details, 9 of 10

Picture No5, grey top, red jeans: again a very good picture of heavy rain, all finger on her right hand, but finger loo very long, an the color of the left hand does not match the rest. 8 of 10

Picture No6: grey top in a room with white tiles: Well, there is no obvious explaination why she is all wet, but that does not matter, the expression of her face shows fun , you can imagine any funny reason , her shirt is all wet, but her jeans looks dry, also no waterdrops on her skin, even so her hair is wet. the level of wetness of shirt and the rest do not match because of that  7 of 10.

Again, my judgement is mainly based on the AI, which the technical aspects.  The photograf (artist) in me might judge a bit different as he also looks at the creativity of the picture, so if you ask which picture i would print and hang to my wall, No1 would clearly win.

Dear Hudson, i really admire your work and skills. please continue, i am looking forward to see more  

By Hudson - Sat 04/01/25 21:29:22 GMT

Website:

First off, thanks to Jake, Quiet271, and Oldbewet who responded to my last post!

Quiet271: Thanks for your questions about my process...I hope you'll understand that I want to keep some of that under wraps for now, mostly because I'm not so confident in the output thus far....I'm really only sharing it because I think people like you interested!

- Lesson one I learned: the focus has to be Flux now. And we still have __many__ limitations with Flux. It still doesn't understand what wet fabric is. Or what it means to sit in a pool with your clothes on. Stable Diffusion will always have a place in my heart though LOL..

- Lesson two: you have to train the models yourself if you want the best results. The other models did a great job, but the results from my own model I think are a little better. But the huge thing is that I get passable results WAY more often. Like 1 out of 2 generations is OK, whereas it used to be like 1 out of 40 with the third-party LoRAs.

Oldbewet: you have an eagle-eye for details. Let me know what you think, if you can, and I'll try to improve the model. I too noticed the bee-stung fingers of one of the models in the last update I made! hahah.

But in a way, sometimes anomalies like that are more real. Less produced and "perfect".

My (the?) problem is that AI still cannot synthesize. The best results come from scenarios on which the model is trained. I think the last photo, of the girl in the bathhouse (?) with the black-and-white shirt probably turned out the best. No surprise why, I trained my model on professional content. Which kinda sucks because the situations are super boring, but usually the photography is the better than candid photos. Ah!! what if I trained on art sessions that weren't from wetlook producers...there's a thought for next time!

I tried to make a pic of a girl in a phone booth, only it's raining inside the phone booth. The models didn't really "get" that.

Thank you all very much!!!!!!!